Fukuoka Institute of Technology REPOSITORY SYSTEM

HOME    repository info    mypage        Japanese    library    university    Feedback

Fukuoka Institute of Technology Repository >
02.University Bulletins & Report >
null >
Vol.1 >

 
Title :Echo state network による時系列からの非定常性の検出
Authors :山口 裕
兒嶋 大也
宮嵜 鋼
Issue Date :Dec-2018
Abstract :We developed a novel method that detecting nonstationarity in observed time-series. Using reservoir computing approach, which now becomes a popular tool in machine learning field, we estimated cross-prediction errors between different segments of time-series. These errors are considered as a measure of non-similarities between different segments. Then, multidimensional scaling was used to obtain a reduced representation of relational patterns among segments.
Type Local :紀要論文
ISSN :24345725
Publisher :福岡工業大学総合研究機構
Comment :情報科学研究所(Computer Science Laboratory)
URI :http://hdl.handle.net/11478/1219
citation :福岡工業大学総合研究機構研究所所報
1
43
46
Citation :福岡工業大学総合研究機構研究所所報 Vol.1 p.43 -46
Appears in Collections:Vol.1

Files in This Item:

File Description SizeFormat
11478-1219_p43.pdf790KbAdobe PDFView/Open